Purpose
This post is related to https://github.com/RobertsLab/resources/issues/2251
Run Chilean mytilus methylation samples via nextflow methylseq.
Files needed:
- zipped fastqs: https://gannet.fish.washington.edu/v1_web/Raw_WGBS_Mch/
- genome: https://gannet.fish.washington.edu/v1_web/owlshell/bu-github/project-chilean-mussel/data/Mchi/
- config file:
- uw_hyak_srlab.config
- This file is specific for this analysis using alignment score of -0.8 and attempting to perform the analysis first on coenv node, then on srlab node, then on ckpt-all because we already know it takes ~6 hours to align each fastq.
- samplesheet: samplesheet.csv
- the format of this samplesheet is standard for nf-core pipelines and is a 3 column csv file that lists sample name, path of fastq1, and path of fastq2. An example of the first two lines is shown below. The format including the header must be exact.
sample,fastq_1,fastq_2
LCo_BSr1,/gscratch/scrubbed/strigg/analyses/20250731_methylseq/raw-reads/LCo_BSr1_R1.fastq.gz,/gscratch/scrubbed/strigg/analyses/20250731_methylseq/raw-reads/LCo_BSr1_R2.fastq.gz
Methods
Pipeline run on 8/5/2025
# start screen session
screen -S methylseq0801
# start interactive node
salloc -A srlab -p cpu-g2-mem2x -N 1 -c 1 --mem=30GB --time=96:00:00
mamba activate nextflow
# zip fastq files in /gscratch/scrubbed/strigg/analyses/20250731_methylseq
gzip *.fastq
# run methylseq pipeline from /gscratch/scrubbed/strigg/analyses/20250731_methylseq
nextflow run nf-core/methylseq \
-c /gscratch/scrubbed/strigg/analyses/20250731_methylseq/uw_hyak_srlab.config \
--input /gscratch/scrubbed/strigg/analyses/20250731_methylseq/samplesheet.csv \
--outdir /gscratch/scrubbed/strigg/analyses/20250731_methylseq \
--fasta /gscratch/scrubbed/strigg/analyses/20250731_methylseq/MchilensisGenomeV1.fa \
--accel \
-resume \
-with-report nf_report.html \
-with-trace \
--nomeseq
rsync --archive --verbose --progress --exclude=work/ --exclude=*.bam --exclude=BismarkInde
x/ --exclude=*.fa --exclude=raw-reads/ 20250731_methylseq shellytrigg@gannet.fish.washington.edu:/volume2/web/metacarcinus/Mchilensis
Results
- multiqc report here: multiqc_report.html
- all pipeline output is here: https://gannet.fish.washington.edu/metacarcinus/Mchilensis/20250731_methylseq/
- bismark results: bismark
the trimming looks like it could be improved for all read 2 based on the multiqc.
Step-by-step instructions for re-running with improved trimming
If you don’t have mamba in your path, run:
/gscratch/srlab/nextflow/bin/miniforge/bin/mamba init
Then run:
source ~/.bashrc
create a working directory and change into it
# this is just an example, and you'd replace what's in '< >' with your username and the date
mkdir /gscratch/scrubbed/<username>/analyses/<date>_methylseq
cd /gscratch/scrubbed/<username>/analyses/<date>_methylseq
create a screen session
screen -S nextflow
request an interactive node. This run should complete in about 24 hours, but requesting 48 hours to be safe. We don’t need a lot of memory here hence 30GB.
salloc -A srlab -p cpu-g2-mem2x -N 1 -c 1 --mem=30GB --time=48:00:00
activate the nextflow environment
mamba activate nextflow
run the methylseq nf-core pipeline with the improved trimming parameter. NOTE: you need to change the outdir to your working directory that you created above. You will also have to change the paths of the .config file, the --input samplesheet.csv, and the --fasta genome.fa if you run this after 8/21/2025 because of the file shelf life on scrubbed.
nextflow run nf-core/methylseq \
-c /gscratch/scrubbed/strigg/analyses/20250731_methylseq/uw_hyak_srlab.config \
--input /gscratch/scrubbed/strigg/analyses/20250731_methylseq/samplesheet.csv \
--outdir /gscratch/scrubbed/<username>/analyses/<date>_methylseq \
--fasta /gscratch/scrubbed/strigg/analyses/20250731_methylseq/MchilensisGenomeV1.fa \
--accel \
--clip_r2 10 \
-resume \
-with-trace \
-with-report nf_report.html \
-with-timeline nf_timeline.html \
--nomeseq